Formation of a hydrogen-bonded heptazine framework by self-assembly of melem into a hexagonal channel structure

Article abstract of DOI:10.1002/chem.201103527

Self-assembly of melem C₆N₇(NH₂) ₃ in hot aqueous solution leads to the formation of hydrogen-bonded, hexagonal rosettes of melem units surrounding infinite channels with a diameter of 8.9 A. The channels are filled with strongly disordered water molecules, which are bound to the melem network through hydrogen bonds. Single-crystals of melem hydrate C₆N₇(NH₂) ₃·xH₂O (x≈2.3) were obtained by hydrothermal treatment of melem at 200 °C and the crystal structure (R βar 3χ, a=2879.0(4), c=664.01(13) pm, V=4766.4(13)× 10⁶ pm³, Z=18) was elucidated by single-crystal X-ray diffraction. With respect to the structural similarity to the well-known adduct between melamine and cyanuric acid, the composition of the obtained product was further analyzed by solid-state NMR spectroscopy. Hydrolysis of melem to cyameluric acid during syntheses at elevated temperatures could thus be ruled out. DTA/TG studies revealed that, during heating of melem hydrate, water molecules can be removed from the channels of the structure to a large extent. The solvent-free framework is stable up to 430 °C without transforming into the denser structure of anhydrous melem. Dehydrated melem hydrate was further characterized by solid-state NMR spectroscopy, powder X-ray diffraction, and sorption measurements to investigate structural changes induced by the removal of water from the channels. During dehydration, the hexagonal, layered arrangement of melem units is maintained whereas the formation of additional hydrogen bonds between melem entities requires the stacking mode of hexagonal layers to be altered. It is assumed that layers are shifted perpendicular to the direction of the channels, thereby making them inaccessible for guest molecules. Copyright

Full text of DOI:10.1002/chem.201103527

DOI: 10.1002/chem.201103527
Formation of a Hydrogen-Bonded Heptazine Framework by Self-Assembly
of Melem into a Hexagonal Channel Structure
Sophia J. Makowski, Pia Kçstler, and Wolfgang Schnick*[a]
Abstract: Self-assembly of melem
C6N7A(NH2)3 in hot aqueous solution
adduct between melamine and cyanuric
acid, the composition of the obtained
product was further analyzed by solid-
state NMR spectroscopy. Hydrolysis of
melem to cyameluric acid during syn-
theses at elevated temperatures could
thus be ruled out. DTA/TG studies re-
vealed that, during heating of melem
hydrate, water molecules can be re-
moved from the channels of the struc-
ture to a large extent. The solvent-free
framework is stable up to 4308C with-
out transforming into the denser struc-
ture of anhydrous melem. Dehydrated
melem hydrate was further character-
ized by solid-state NMR spectroscopy,
powder X-ray diffraction, and sorption
measurements to investigate structural
changes induced by the removal of
water from the channels. During dehy-
dration, the hexagonal, layered ar-
rangement of melem units is main-
tained whereas the formation of addi-
tional hydrogen bonds between melem
entities requires the stacking mode of
hexagonal layers to be altered. It is as-
sumed that layers are shifted perpen-
dicular to the direction of the channels,
thereby making them inaccessible for
guest molecules.
CHTUNGTRENNUNG
leads to the formation of hydrogen-
bonded, hexagonal rosettes of melem
units surrounding infinite channels with
a diameter of 8.9 ꢀ. The channels are
filled with strongly disordered water
molecules, which are bound to the
melem network through hydrogen
bonds. Single-crystals of melem hydrate
C6N
G
tained by hydrothermal treatment of
melem at 2008C and the crystal struc-
¯
ture (R 3c, a=2879.0(4), c=664.01(13)
Keywords: crystal structure · hy-
drogen bonds · self-assembly · sol-
id-state structures · stacking interac-
tions
pm, V=4766.4(13)ꢁ106 pm3, Z=18)
was elucidated by single-crystal X-ray
diffraction. With respect to the struc-
tural similarity to the well-known
Introduction
the formation of a channel structure is the 1:1 adduct be-
tween the s-triazine (C3N3) compounds cyanuric acid and
melamine (CA·M) (see Scheme 1).[2,4,5,6] In this adduct, cya-
nuric acid and melamine form hydrogen-bonded rosette-like
Self-assembly, that is, the spontaneous arrangement of mole-
cules into ordered structures by weak, reversible interac-
tions, plays not only an important role in biological systems,
for example, for protein folding or pairing of nucleotides,
but has also attracted considerable interest in synthetic and
materials chemistry.[1,2] In supramolecular chemistry or crys-
tal engineering, complex chemical structures determined by
several types of noncovalent interactions such as hydrogen
bonds, ionic, hydrophobic, van der Waals, or dispersive
forces are formed. Next to the ambition to understand the
principles of molecular self-assembly in the solid state, the
synthesis of solids that exhibit desired structural motifs such
as ordered channels or cavities reminiscent of zeolites, is a
relevant goal in crystal engineering.[3] An outstanding exam-
ple for the noncovalent assembly of molecules resulting in
[a] S. J. Makowski, P. Kçstler, Prof. Dr. W. Schnick
Department Chemie
Lehrstuhl fꢂr Anorganische Festkçrperchemie
Ludwig-Maximilians-Universitꢃt Mꢂnchen
Butenandtstrasse 5-13 (D)
81377 Mꢂnchen (Germany)
Fax : (+49)89-2180-77440
E-mail: wolfgang.schnick@uni-muenchen.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103527.
Scheme 1. Hydrogen-bonded rosettes formed between the s-triazine com-
pounds isocyanuric acid and melamine.
3248
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3248 – 3257

FULL PAPER
hexamers that are arranged in planar sheets. Similar rosette-
like motifs have been found in adducts between trithiocya-
nuric acid and melamine[6] or between trithiocyanuric acid
and 4,4ꢅ-bipyridyl.[7] Furthermore, self-assembly of mole-
cules into hydrogen-bonded rosettes has also been reported
for single-component systems, for example, derivatives of
trimesic acid, trimesic amide, isophthalic acid, or orthoboric
acid.[8,9] However, particular stacking of these hexamers re-
sulting in the formation of channels could be observed only
in a few cases.[9] Among these examples of hydrogen-
bonded rosette motifs, the CA·M adduct is certainly the best
studied. Whitesides and co-workers have not only investigat-
ed in detail the self-assembly of cyanuric acid and melamine
into the respective rosette structures but have also devel-
oped concepts for the directed construction of such rosettes.
By employing s-triazine molecules with sterically bulky sub-
stituents or allowing preorganization of building units by in-
terconnecting them with a hub linker, the formation of ro-
settes can be favored over the formation of tapes.[2,5] How-
ever, whereas numerous compounds based on the CA·M lat-
tice have been extensively studied by many groups, no com-
parable structures comprising the analogous but larger
s-heptazine (C6N7) compounds cyameluric acid or melem
(see Scheme 2) are known so far. The molecular structure of
and in extended CNx polymers,[15] as well as quantum chemi-
cal calculations that suggest s-heptazine as likely building
[16,17]
block of graphitic C3N4
have contributed to increased
interest in s-heptazines. Numerous substituted s-heptazines,
for example, cyameluric chloride C6N7Cl3,[16,18] cyameluric
7ACTHNUTRGNEUNG
azide C6N7(N3)3,[19] melonates (C6N (NCN)3)3À,[20,21] and sev-
eral organic derivatives[22,23] have been synthesized and in-
vestigated in detail. The structures of melem[24,25] and cya-
meluric acid[26,27] as well as of numerous salts[22,28,29,30] have
also been solved recently. The employment of melemium
and cyamelurate salts has opened access to the chemistry of
s-heptazines in solution. In addition, the thermal condensa-
tion process of melamine leading to melem has been eluci-
dated. Contrary to former assumptions that the condensa-
tion might proceed via the intermediate melam (C3N3-
AHCTNUGTERNG(NNU NH2)2)2NH it could be shown that during thermal treat-
ment of melamine, crystalline melem is formed via several
intermediate melamine–melem adduct phases.[31,32] In the
eventually obtained product, melem molecules are arranged
in corrugated layers and are interconnected by a dense net
of hydrogen bonds both within and between the layers.[24,25]
The importance of intermolecular hydrogen bonds for the
arrangement of melem units has been demonstrated not
only for the assembly of melem molecules in the crystal but
also for monolayers of melem on AgACTHNUTRGNEUNG(111), as has been in-
vestigated by scanning tunneling microscopy studies recent-
ly.[33]
In this contribution, we report on the self-assembly of
melem in aqueous solution, thus providing the first example
of a structure with a hydrogen-bonded rosette-like pattern,
comparable to the CA·M adduct, that comprises s-heptazine
units. The importance of solvent molecules for the formation
of such rosettes and of hexagonal channels in this compound
in comparison to structures obtained under anhydrous con-
ditions is exposed. Furthermore, we demonstrate the
strength of intermolecular hydrogen bonds in this material,
which stabilize the hexagonal arrangement of melem units
even in the dehydrated material.
Scheme 2. The s-triazine compounds melamine A and isocyanuric acid B,
and the respective s-heptazine compounds melem C and isocyameluric
acid D.
the s-heptazine or cyameluric ring system, respectively, has
already been postulated by Pauling and Sturdivant in
1937.[10] However, it was not proven until almost fifty years
later when the crystal structure of molecular s-heptazine
C6N7H3 could be solved.[11] Due to the low solubility of s-
heptazines in common solvents, this building unit is consid-
erably less frequently employed in chemical syntheses than
s-triazine compounds. However, in the last two decades
carbon nitride type compounds have attracted considerable
interest as materials for numerous technical applications
due to the high thermal and oxidation stability of this class
of compounds as well as their semiconducting and catalytic
properties.[12] This revival in research of molecular and poly-
meric carbon nitride type compounds has also initiated sub-
stantial progress in the field of s-heptazine chemistry. Fur-
thermore, the identification of s-heptazine as a building
block in the polymers melon and poly(heptazine imide)[13,14]
Results and Discussion
Synthesis and crystal structure: The s-heptazine compound
melem 1 is commonly synthesized by thermal treatment of
less condensed C/N/H compounds such as melamine
C3N3ACHTUNGTRNEUNG(NH2)3 or dicyandiamide (NH2)2CN(CN), leading to
the elimination of ammonia and subsequent condensation.
Crystalline melem of high purity is prepared by heating
small amounts of starting materials in sealed ampoules.[24]
An alternative approach for the synthesis of larger quanti-
ties of melem by heating melamine in open systems has
been reported recently.[30,34] In this procedure, a reaction
mixture of melem and unreacted melamine is obtained.
Final purification by boiling the product in water to remove
unreacted melamine yields melem as a hydrate 2, which has
not been structurally characterized so far.[30,34] By reacting
melem under hydrothermal conditions in an autoclave at
Chem. Eur. J. 2012, 18, 3248 – 3257
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3249

W. Schnick et al.
Table 1. Crystallographic data and details of the structure refinement for
C6N7A(NH2)3·1.98H2O.
agreement with the less dense packing of melem molecules
in melem hydrate 2 compared with anhydrous melem 1, in
anhydrous melem, each molecule is connected to five other
melem molecules each through two hydrogen bonds;[24,25] in
melem hydrate, hydrogen bonds are formed only to four fur-
ther molecules (see Figure 3 and Figure 4). For hydrogen
bonds, it is a common approach to estimate their approxi-
mate strength based on distances and angles between
CHTUNGTRENNUNG
C6N7ACHTNUGTRNE(UGN NH2)3·1.98H2O
molar mass [gmolÀ1
crystal system
space group
T [K]
]
252.85
trigonal
¯
R 3c (no. 167)
173
diffractometer
radiation, l [pm]
a [pm]
Nonius Kappa-CCD
Mo Ka, 71.073
2879.0(4)
À
donors and acceptors. All intermolecular N H···N interac-
c [pm]
664.01(13)
4766.4(13)
18
1.585
0.24ꢁ0.04ꢁ0.02
0.126
tions between melem units can therefore be classified as
medium to strong (see Table 2).[35] As can be expected due
to the presence of channels in the structure, the water mole-
cules are strongly disordered and are localized on partially
occupied sites. Refinement of oxygen occupancies in the
single-crystal data of 2 yielded the formula melem·1.98H2O.
The amount of crystal water was further determined by
means of elemental analysis (melem·2.50H2O) and thermal
analysis (melem·2.33H2O), resulting in an average formula
V [106 pm3]
Z
calculated density [gcmÀ3
crystal size [mm3]
]
absorption coefficient [mmÀ1
diffraction range
]
3.728ꢁqꢁ27.388
À37ꢁhꢁ37, À31ꢁkꢁ31,
À8ꢁlꢁ8
index range
parameters/restraints
total no. of reflections
no. of independent refl.
no. of observed refl.
109/2
4690
1218
769
C6N7ACHTUNGTRENUNG(NH2)3·2.3 H2O for melem hydrate 2. Due to the strong
disorder of water molecules in the channels, not all hydro-
gen positions could be determined unambiguously from the
single-crystal data and were defined by considering reasona-
min./max. residual electron density
À0.282/+0.619
[eꢁ10À6 pmÀ3
GoF
]
1.047
final R indices [I>2s(I)][a]
R1=0.0624, wR2=0.1633
R1=0.1083, wR2=0.1892
À
À
À
ble O H distances and meaningful O H···O and O H···N
hydrogen-bonding interactions (see Figure 3). Interestingly,
the above described assembly of melem molecules in hydro-
gen-bonded rosettes appears to be a preferred arrangement
for melem. Scanning tunneling microscopy studies of melem
final R indices (all data)[a]
2
[a] w=[s
N
2008C, we were able to obtain single-crystals of this com-
pound and determine its crystal structure. Crystallographic
data for melem hydrate 2 and details of the structure refine-
ment are summarized in Table 1.
on AgACTHNGUTERN(UNG 111) revealed, among several other polymorphs of
melem monolayers, also hydrogen-bonded hexamers of
melem;[33] and in the adduct phase 3 melem·melamine 3-
ACHNUTTGREG(NNUN C6N7ACHTUTNGERNNUG(NH2)3)·ACHUTNEGTRG(NNUN C3N3ACHTNUGTRENUN(NG NH2)3), the arrangement of melem mol-
Treatment of melem in boiling water breaks up the inter-
molecular hydrogen bonds and leads to a rearrangement of
molecules into a hydrogen-bonded rosette-like network (see
Figure 1, bottom right). During the self-assembly of melem
in solution, water molecules act as a template, resulting in
the formation of a porous structure in contrast to the denser
packed structure of anhydrous melem. The rosettes sur-
round hexagonal channels with a diameter of 8.9 ꢀ (van der
Waals radii subtracted) that are filled with water molecules
(see Figure 1, top). The self-assembly of molecules in ro-
settes resulting in the formation of channels resembles the
structure of a famous case of noncovalent synthesis, namely
the adduct between cyanuric acid and melamine (CA·M).
This s-triazine (C3N3) based compound exhibits cavity diam-
eters of approximately 4 ꢀ;[6] the use of the larger s-hepta-
zine (C6N7) compound melem therefore leads to a remark-
able increase in channel diameter by a factor of 2.2. The
melem units are not coplanar but tilted out of the ab plane
and are arranged in a helix motif along the crystallographic
c axis (see Figure 1, bottom left) resulting in a corrugated
layered structure perpendicular to the channels (see
Figure 2). The interlayer distance is 337 pm and, hence, is
slightly larger than in anhydrous melem 1 (327 pm). For the
observed assembly of melem units in the structure of melem
hydrate, the formation of a dense network of hydrogen
bonds appears to be the determining factor (see Table 2). In
ecules resembles that of melem hydrate 2. Whereas the
melem rosettes are filled with water in melem hydrate 2, in
the adduct phase (3 melem·melamine), a melamine molecule
is located in the middle of these voids.[32]
Solid-state NMR spectroscopy: Because the self-assembly
motif of melem hydrate 2 resembles that of the well-known
adduct between cyanuric acid and melamine (CA·M), we
suspected that, actually, not melem but an analogous adduct
between the s-heptazine compounds cyameluric acid
(C6N7O3H3) and melem (C6N7ACHTUNTRGNEUNG(NH2)3) might have been
formed due to partial hydrolysis of melem. Because the OH
and NH2 groups cannot be unequivocally differentiated by
X-ray diffraction, we examined the obtained product by
means of solid-state NMR spectroscopy. The observed sig-
nals could be assigned in accordance with solid-state NMR
studies of anhydrous melem 1[24] (compare Scheme 3). The
13C NMR signals of CN
served at d=163.5 and 153.2 ppm, respectively (see
Figure 5, top). For both the CN3 and CN2A(NH2) carbon
2ACHTUGNTRENN(GNU NH2) and CN3 groups were ob-
CTHUNGTRENNUNG
atoms there are each two crystallographic positions but be-
cause the chemical surroundings of these atoms basically
only differ with respect to hydrogen-bonding interactions of
adjacent nitrogen atoms, the respective resonances are not
distinctly split. In the 15N NMR spectrum (Figure 5, bottom)
signals can be attributed to the amino groups (d=
3250
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3248 – 3257

Formation of a Heptazine Framework
FULL PAPER
À272.7 ppm), the central nitro-
gen atom of the s-heptazine
core (d=À236.2 ppm), and the
remaining nitrogen atoms of
the
ring
(d=À206.9
to
À201.3 ppm). Moreover, the
multiplicity of signals in the
15N NMR spectrum is in ac-
cordance with the number of
crystallographic nitrogen posi-
tions. Considering the number
of signals in the NMR spectra,
the presence of an adduct
phase between melem and cya-
meluric acid can be ruled out.
Such an adduct phase would
lead to splitting of signals both
for the outer nitrogen atoms of
the ring N(o) and the carbon
atoms. Especially for the 13C
NMR signal at d=163.5 ppm,
which corresponds to carbon
atoms that are bound either to
NH2 or OH substituents,
a
splitting would be expected.
Elemental analysis and mass
spectrometry
further
con-
firmed that detectable amounts
of cyameluric acid are not
formed during the treatment of
melem in boiling water. Ele-
mental analysis revealed an
atomic
matching
ratio
the
C/N=12:20,
formula
C6N
(NH2)3,
whereas
an
Figure 1. Crystal structure of melem hydrate 2. (Top) Representation of the porous structure with hexagonal
channels along c, filled with water molecules; (Bottom left) Arrangement of melem molecules in helices along
c; (Bottom right) Rosette formed of six melem molecules with a diameter of 8.9 ꢀ (van der Waals radii sub-
tracted), filled with water; dashed lines represent hydrogen bonds.
adduct between melem and cy-
ameluric acid would corre-
spond to a ratio of 12:17. Con-
sistently, an intense peak at m/
z=218.2 (C6N10H6) occurs in
the mass spectrum, but none at
m/z=221.1 (C6N7O3H3) (see the Supporting Information,
Figure S1–S2).
Furthermore, a dehydrated sample of melem hydrate has
been analyzed by solid-state NMR spectroscopy. According
to TG/DTA data (see below), crystal water can be largely
removed by heating the compound to 1508C under vacuum
overnight. In comparison to as-synthesized melem hydrate,
Table 2. Donor–acceptor distances and donor–hydrogen–acceptor angles
(pm/8) for the hydrogen-bonding network in C6N7ACHTNUTRGNEU(GN NH2)3·1.98H2O.
À
À
D H···A
D···A
<D H···A
N5-H1···N4
N6-H2···N3
N6-H3···O1
O1-H4···O1
O1-H5···O2
304.0
294.8
291.6
348.3
277.9
176.10
173.70
130.92
139.43
134.37
Figure 2. Layered structure of melem hydrate 2. Arrows indicate the po-
sitions of hexagonal channels.
Chem. Eur. J. 2012, 18, 3248 – 3257
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3251

W. Schnick et al.
Scheme 3. 13C and 15N NMR chemical shifts observed for melem hydrate
2.
Figure 3. Hydrogen bonds in melem hydrate 2 between melem units
(left) and between the melem network and water molecules (right).
Figure 5. 13C (top) and 15N (bottom) CP-MAS solid-state NMR spectra of
as-synthesized melem hydrate 2, dried at room temperature.
Figure 4. Crystal structure of anhydrous melem 1. Illustration of the lay-
ered structure (top); hydrogen-bonding interactions between melem
units (bottom). Structural data for anhydrous melem 1 were taken from
the literature.[25]
whereas the multiplicity of the remaining signals do not
change. Hence, the chemical surroundings of the s-heptazine
cores remain basically unaltered during the removal of crys-
tal water but those of the amino groups, and, consequently,
also of the carbon atoms they are bound to, are different for
as-synthesized and dehydrated melem hydrate. Removal of
water from the channels corresponds to a withdrawal of hy-
drogen-bonding donors and acceptors from the structure.
Consequently, dehydration causes melem units to form addi-
both 13C and 15N NMR spectra of the dehydrated sample
(see Figure 6) show a broadening of signals, as expected due
to the lower crystallinity of this material. Additionally, split-
ting of the CN
2ACHTUNGTRENNUNG
(NH2) signal in the 13C NMR spectrum and
the NH2 signal in the 15N NMR spectrum are observed,
3252
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3248 – 3257

Formation of a Heptazine Framework
FULL PAPER
Figure 7. TG (solid) and DTA (dashed) curves of melem hydrate 2
(22.4 mg), measured with a heating rate of 5 KminÀ1
.
densation to a melon-like product.[14] Above 5858C, the
rapid mass loss of 70.1% accompanied by a strong endother-
mic signal in the DTA curve indicates complete decomposi-
tion of the sample into gaseous products. The incipient re-
lease of crystal water already at 208C is consistent with the
above results from single-crystal data, indicating that partly,
only weak hydrogen bonds exist between the melem frame-
work and water molecules. During temperature-dependent
powder X-ray diffraction (see the Supporting Information,
Figure S3) between room temperature and 5008C, no emer-
gence of any reflections corresponding to crystalline anhy-
drous melem 1 could be observed. Hence, melem hydrate 2
starts to decompose above 4308C without previous phase
transition to crystalline melem 1.
Figure 6. 13C (top) and 15N (bottom) CP-MAS solid-state NMR spectra of
melem hydrate 2, dehydrated at 1508C under vacuum.
In agreement with TG/DTA measurements, melem hy-
drate can be largely dehydrated by heating to 1508C under
vacuum. However, elemental analysis of a dehydrated
sample yielded a total amount of nitrogen, carbon and hy-
drogen of only 97.8 wt.%. Assignment of the remaining 2.2
wt.% to oxygen results in the formula C6.0N9.9O0.3H6.2. This
composition can be attributed either to incomplete dehydra-
tional hydrogen bonds to other melem entities. The amino
À
groups are affected most by dehydration because N H···O
interactions between the melem framework and crystal
water molecules are primarily formed by them (see
Figure 3). Therefore, changes in the NMR spectra are
mainly observed for the amino groups. Considering the geo-
metric prerequisites for a stable hydrogen-bonding network,
tion of melem hydrate [ca. C6N
7ACTHNUGTREN(NUNG NH2)3·0.3H2O] or to partial
À
formation of additional N H···N interactions is likely to re-
quire a shifting of melem units as is discussed in detail
below.
hydrolysis [ca. (C6N7A(NH2)2.7(OH)0.3)] due to reaction of
CHTUNGTRENNUNG
melem with crystal water at elevated temperatures. Howev-
er, because such a small degree of hydrolysis cannot be
clearly detected by solid-state NMR or IR spectroscopic
analyses, none of these possibilities can be ruled out unam-
biguously. A comparison of the powder X-ray diffraction
patterns of as-synthesized and dehydrated melem hydrate
shows that, according to expectation, removal of crystal
water significantly decreases the crystallinity of the product
(see Figure 8, top). However, the sample does not become
completely amorphous; the main reflections in the X-ray
diffraction pattern are retained even after dehydration of
melem hydrate. Especially, the 110 and 220 reflections, cor-
responding to a hexagonal pattern, are clearly visible (see
Figure 8, bottom). Furthermore, a broad reflection at 2q=
26–288 is observed that can be attributed to a layered ar-
Thermal behavior: The strength of interactions between
melem and water molecules as well as the thermal stability
of melem hydrate 2 was investigated by TG and DTA mea-
ACHTUNGTRENNUNGsurements. The obtained curves (see Figure 7) show that de-
hydration of 2 takes place in two steps at 20–1508C (mass
loss observed: 16.1%, calculated for melem·2.33H2O:
16.1%). Subsequently, the dehydrated sample shows ther-
mal stability up to 4308C. Above this temperature the ther-
mal behavior of 2 is as typically observed for molecular
s-heptazine compounds in this temperature range.[20,24,26,29]
The slow mass loss of 11.4% at 430–5858C can presumably
be assigned to the release of ammonia and subsequent con-
Chem. Eur. J. 2012, 18, 3248 – 3257
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3253

W. Schnick et al.
cules can essentially be regarded as a template that is re-
quired during self-assembly of melem molecules to obtain
the porous structure of 2 and not the denser structure of 1.
Melem hydrate, therefore, not only represents an interesting
example for the development of a highly symmetric channel
structure by self-assembly but also a thermally stable s-hep-
tazine-based framework that is predominantly stabilized by
strong intermolecular hydrogen bonds between the melem
units.
Sorption measurements: To investigate whether removal of
crystal water makes the channels accessible for other guest
molecules, N2 gas adsorption measurements were per-
formed. After dehydrating a sample of melem hydrate 2 for
12 h at 208C and subsequently for 6 h at 1508C, an isotherm
was recorded between P/Po =0.02 and P/Po =0.5. Applying
the Brunauer–Emmett–Teller (BET) model to the isotherm
resulted in a surface area SBET =29 m2 gÀ1. In comparison to
2D covalent organic frameworks (COFs) with similar pore
diameters that exhibit surface areas of 711 m2 gÀ1 (COF-1)
or 980 m2 gÀ1 (COF-6), respectively,[36] it becomes evident
that, for dehydrate melem hydrate 2, perceptible filling of
channels with nitrogen does not occur. As discussed above,
this might be due to a shifting of melem layers perpendicu-
lar to the direction of the channels, thereby locking the
channels and making them inaccessible for guest molecules.
FTIR spectroscopy: Melem hydrate 2 was further analyzed
by FTIR spectroscopy (Figure 9). In the spectra of both an
as-synthesized sample dried at room temperature and of a
sample dehydrated at 1508C, all signals of bending and
stretching vibrations of the s-heptazine core are in accord-
ance with data of anhydrous melem 1.[24] In the region of
NH and OH stretching vibrations at 2700–3550 cmÀ1, sharp
n(NH) bands, which have been observed for anhydrous
melem 1, are covered by a broad n(OH) signal in the spec-
trum of as-synthesized melem hydrate but are observable in
the spectrum of dehydrated melem hydrate.
Figure 8. Powder X-ray diffraction patterns (293 K, l=154.06 pm) of
melem hydrate 2 (top); a) calculated pattern based on single-crystal data;
b) measured pattern of an as-synthesized sample, dried at room tempera-
ture, and c) measured pattern of a sample dehydrated at 1508C under
vacuum. Representation of distances in the structure of melem hydrate
corresponding to 110 and 220 reflections (bottom).
rangement of melem molecules with an interlayer distance
of 320–340 ppm. However, there is no evidence that the
stacking pattern of hexagonal layers of melem units also re-
mains unaltered. Sorption measurements (see below) indi-
cate that during dehydration of melem hydrate, channels in
the structure are blocked. This might be explained by a
shifting of layers perpendicular to the c-axis. As discussed
above, removal of water causes the amino groups of melem
to form additional hydrogen bonds to other melem units.
Due to perpetuation of the hexagonal arrangement within
the layers, these newly formed hydrogen bonds are most
likely formed between the layers. Shifting of the layers
might be required so that the amino groups can serve most
suitably as donors and acceptors, respectively, for hydrogen
bonds.
In summary, the hexagonal structure of melem hydrate
does not collapse when water is removed from the channels
but the stacking mode of melem layers is probably altered.
Because the hydrogen-bonding network interconnecting the
melem molecules appears to be strong enough to preserve
the main structural features during dehydration, water mole-
Figure 9. FTIR spectra of a) as-synthesized melem hydrate 2, dried at
room temperature and b) melem hydrate 2, dehydrated at 1508C under
vacuum.
3254
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3248 – 3257

Formation of a Heptazine Framework
FULL PAPER
(øext. =16 mm, øint. =12 mm). The tube was filled with argon, sealed at a
length of 16 cm and placed in an inclined tube furnace. The sample was
heated at 1 KminÀ1 to 4508C, held at this temperature for 5 h and cooled
to RT at 2 KminÀ1. From the bottom part of the ampoule, crystalline
melem (104 mg, 0.47 mmol, 60%) could be isolated. Elemental analysis
Conclusion
In this contribution we describe the self-assembly of melem
in aqueous solution, thereby complementing preceding stud-
ies on the arrangement of melem units in the solid state and
calcd (%) for C6N
2.78, N 63.81.
7ACHTUNGERTN(NUNG NH2)3: C 33.03, H 2.78, N 64.21; found: C 33.17, H
in monolayers on AgACTHNUTRGNE(NUG 111). Both in anhydrous melem and in
Single crystals of melem hydrate suitable for X-ray diffraction analysis
were obtained under hydrothermal conditions. A suspension of raw
melem (150.0 mg, 0.69 mmol) in H2O (20 mL) acidified with a few drops
of 2m HCl was filled in a Teflon-lined 40 mL steel autoclave, heated at
melem hydrate, the formation of numerous intermolecular
hydrogen bonds appears to be the main factor determining
the development of a specific structure. Although melem is
insoluble in aqueous solution, treatment of melem in boiling
water induces rearrangement of melem units from a dense
to a porous structure. In the presence of solvent molecules,
hydrogen bonds in melem are broken up and, during self-as-
0.07 KminÀ1 to 2008C and cooled to RT at 0.25 KminÀ1
.
For the preparation of larger quantities of melem hydrate, crystalline
melem was used as starting material because, in bulk samples of melem
hydrate synthesized from raw melem, small amounts of amorphous side
phases could be identified by solid-state NMR spectroscopy. Melem was
boiled in water at reflux temperature for 30 min, filtered, washed with
water and dried at RT, yielding melem hydrate 2 as a colorless powder.
À
sembly into a porous structure, N H···O interactions be-
tween melem and solvent molecules are formed whereas the
(NH2)3+]; elemental analysis calcd
7ACHTUNGTRENNUNG
À
number of N H···N bonds between melem units is de-
MS (DEI+): m/z (%): 218 (100) [C6N
(%) for C6N7A(NH2)3·2.5H2O: C 27.38, H 4.21, N 53.21; found: C 27.52, H
CTHUNGTRENNUNG
creased in comparison to anhydrous melem. Thermal analy-
sis revealed varying degrees of binding strengths for hydro-
gen bonds present in melem hydrate. Whereas the main part
of water can be removed from the material quite easily,
small amounts of crystal water remain in the material even
at elevated temperatures. Dehydration affects the stacking
pattern of melem units but not the hexagonal assembly
within the layers. The observation that the hexagonal struc-
ture is maintained even when solvent molecules are largely
removed demonstrates the potential of a dense network of
intermolecular hydrogen bonds between melem units to sta-
bilize a specific arrangement.
4.02, N 53.00.
Dehydration of melem hydrate was performed by heating the compound
to 1508C under vacuum (1·10À3 mbar) overnight. Elemental analysis
calcd (%) for C6N7ACTHNURGTNEUNG(NH2)3·0.3H2O: C 32.23, H 2.98, N 62.64; found: C
32.48, H 2.83, N 62.45.
X-ray structure determination: Single-crystal X-ray diffraction data of 2
were collected at 173 K with a Kappa CCD diffractometer (Mo Ka radia-
tion, l=71.073 pm). The diffraction intensities were scaled using the
SCALEPACK[37] software package and no additional absorption correc-
tion was applied. The crystal structure was solved by direct methods
(SHELXS-97) and refined against F2 by applying the full-matrix least-
squares method (SHELXL-97).[38] Hydrogen positions were determined
from difference Fourier syntheses considering reasonable oxygen–hydro-
gen distances and were refined isotropically. All non-hydrogen atoms
were refined anisotropically.
The characterization of melem hydrate not only allows
for a better understanding of the role of hydrogen-bonding
in C/N/H compounds but also establishes a further relation-
ship between the main building blocks in carbon nitride
chemistry, namely s-triazines and s-heptazines. The hydro-
gen-bonded, rosette-like pattern in s-heptazine melem hy-
drate resembles that of the famous adduct between the s-tri-
azine compounds cyanuric acid and melamine. However, be-
cause until now a rosette-like assembly could be realized for
melem but not for melamine, and for an adduct between cy-
anuric acid and melamine but not for an analogous adduct
between cyameluric acid and melem, it becomes evident
that the formation of such structures depends on diverse fac-
tors. We have demonstrated that the construction of a sym-
metric framework by self-assembly is not limited to the
well-known CA·M adduct but that it is also possible for an
s-heptazine compound. The presented results will certainly
initiate further synthetic efforts to develop novel hydrogen-
bonded networks of various C/N/H compounds or mixtures
thereof.
Further details of the crystal structure investigation(s) can be obtained
from the Fachinformationszentrum Karlsruhe, 76344 Eggenstein-Leo-
poldshafen, Germany (fax: (+49)7247-808-666; e-mail: crysdata@fiz-
karlsruhe.de,
http://www.fiz-karlsruhe.de/request_for_deposited_
data.html) on quoting the depository number CSD-423755.
Powder X-ray diffraction data under ambient conditions were collected
in transmission geometry with a Huber Guinier G670 Imaging Plate dif-
fractometer using Ge
154.06 pm).
ACHTUNGTERN(NGNU 111)-monochromated Cu-Ka1 radiation (l=
High-temperature in situ X-ray diffraction measurements were per-
formed with a STOE Stadi P powder diffractometer with an integrated
furnace, using GeACHTUNRGTNE(NUG 111)-monochromated Mo Ka1 radiation (l=
70.093 pm). The sample was heated in an unsealed Duran capillary (Ø=
0.3 mm) from RT to 5008C with a ramp of 2 KminÀ1 and scans were col-
lected in steps of 25 K up to 1508C and in steps of 50 K at 150–5008C.
1
Solid-state NMR spectroscopy: H, 13C and 15N MAS NMR spectra were
recorded at RT with a Bruker Avance III-500 spectrometer in an external
magnetic field of 11.7 T. The measurements were performed with a rota-
tion frequency of 10 kHz using 4 mm standard double-resonance MAS
probes (Bruker), which were operated with zirconia rotors. For all experi-
ments, broadband proton decoupling was applied. Both 13C and 15N
nuclei were excited via the proton bath by employing a ramped cross-po-
larization sequence in which the 1H pulse amplitude was decreased line-
arly by 50%. Trimethylsilane (TMS) (13C, 1H) and nitromethane (15N)
were used as references, respectively.
Experimental Section
Thermal analysis: Thermoanalytical measurements were performed with
a Thermoanalyzer TG-DTA92 (Setaram). The sample was heated under
inert atmosphere (He) in an alumina crucible from RT to 7508C with a
Synthesis: Melem hydrate 2 was prepared in preparative amounts accord-
ing to the procedure described by Sattler et al.[30] starting either from raw
or pure crystalline melem. Raw melem was obtained by heating mela-
mine (Fluka, p.a.) in a porcelain crucible covered with a lid in a muffle
furnace at 4258C overnight.[30] To synthesized pure melem,[24] melamine
(200 mg, 1.58 mmol, Fluka, p. a.) was placed in a dried Duran glass tube
heating rate of 5 KminÀ1
.
Sorption measurements: An N2 adsorption isotherm was recorded at
77 K with an Autosorb iQ Instrument (Quantachrome Instruments).
Before measuring the isotherm, the sample was dehydrated under dy-
Chem. Eur. J. 2012, 18, 3248 – 3257
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3255

W. Schnick et al.
namic vacuum for 12 h at 258C and subsequently for 6 h at 1508C. The
Brunauer–Emmett–Teller (BET) model was applied to the isotherm be-
tween P/Po =0.02 and P/Po =0.5.
Wilson, C. J. Loweth, S. C. Zimmerman, Angew. Chem. 1996, 107,
2873–2876; Angew. Chem. Int. Ed. Engl. 1996, 34, 2654–2657;
d) S. V. Kolotuchin, P. A. Thiessen, E. E. Fenlon, S. R. Wilson, C. J.
Loweth, S. C. Zimmerman, Chem. Eur. J. 1999, 5, 2537–2547.
[9] a) A. R. A. Palmans, J. A. J. M. Vekemans, H. Kooijman, A. L.
Spek, E. W. Meijer, Chem. Commun. 1997, 2247–2248; b) R. R.
Shuvalov, P. C. Burns, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 2003, 59, i47-i49; c) J. Yang, J.-L. Marendaz, S. J. Geib,
A. D. Hamilton, Tetrahedron Lett. 1994, 35, 3665–3668; d) W. H.
Zachariasen, Acta Crystallogr. 1954, 7, 305–310.
General techniques: FTIR spectra were recorded under ambient condi-
tions with a Bruker IFS 66v/S spectrometer between 400 and 4000 cmÀ1
.
The samples were measured as KBr pellets in an evacuated cell. Direct
Electron Ionization (DEI+) (70 eV) mass spectra were measured with a
JEOL JMS-700 spectrometer. Elemental analyses for C, H and N were
conducted with elemental analyzer systems Vario EL and Vario Micro
(Elementar Analysensysteme GmbH).
[10] L. Pauling, J. H. Sturdivant, Proc. Natl. Acad. Sci. USA 1937, 23,
615–620.
[11] a) R. S. Hosmane, M. A. Rossman, N. J. Leonard, J. Am. Chem. Soc.
1982, 104, 5497–5499; b) M. Shahbaz, S. Urano, P. R. LeBreton,
M. A. Rossman, R. S. Hosmane, N. J. Leonard, J. Am. Chem. Soc.
1984, 106, 2805–2811.
Acknowledgements
[12] a) F. Goettmann, A. Fischer, M. Antonietti, A. Thomas, Angew.
Chem. 2006, 118, 4579–4583; Angew. Chem. Int. Ed. 2006, 45, 4467–
4471; b) F. Goettmann, A. Thomas, M. Antonietti, Angew. Chem.
2007, 119, 2773–2776; Angew. Chem. Int. Ed. 2007, 46, 2717–2720;
c) H. Grueger, C. Kunath, European Patent EP1436607; d) H. Sugi-
mura, Y. Sato, N. Tajima, O. Takai, Surf. Coat. Technol. 2001, 142–
144, 714–718; e) A. Thomas, A. Fischer, F. Goettmann, M. Anto-
nietti, J.-O. Mꢂller, R. Schlçgl, J. M. Carlsson, J. Mater. Chem. 2008,
18, 4893–4908; f) X. Wang, K. Maeda, A. Thomas, K. Takanabe, G.
Xin, J. M. Carlsson, K. Domen, M. Antonietti, Nat. Mater. 2009, 8,
76–80; g) Z. B. Zhou, R. Q. Cui, Q. J. Pang, G. M. Hadi, Z. M.
Ding, W. Y. Li, Solar Energy Mater. Solar Cells 2002, 70, 487–493.
[13] M. Dçblinger, B. V. Lotsch, J. Wack, J. Thun, J. Senker, W. Schnick,
Chem. Commun. 2009, 1541–1543.
[14] B. V. Lotsch, M. Dçblinger, J. Sehnert, L. Seyfarth, J. Senker, O.
Oeckler, W. Schnick, Chem. Eur. J. 2007, 13, 4969–4980.
[15] a) B. V. Lotsch, W. Schnick, Chem. Mater. 2006, 18, 1891–1900;
b) D. R. Miller, J. R. Holst, E. G. Gillan, Inorg. Chem. 2007, 46,
2767–2774.
[16] E. Kroke, M. Schwarz, E. Horath-Bordon, P. Kroll, B. Noll, A. D.
Norman, New J. Chem. 2002, 26, 508–512.
[17] J. Sehnert, K. Baerwinkel, J. Senker, J. Phys. Chem. B 2007, 111,
10671–10680.
The authors thank the Deutsche Forschungsgemeinschaft DFG (project
SCHN 377/15) and the Fonds der Chemischen Industrie (FCI) for finan-
cial support, and the following people for conducting the physical mea-
ACHTUNGTRENNUNGsurements (all Department Chemie, LMU Mꢂnchen): Brigitte Breiten-
stein (mass spectrometry), Stephan Hug (sorption measurements), Dr.
Peter Mayer (single-crystal data collection), Christian Minke (solid-state
NMR spectroscopy), and Thomas Miller (temperature-dependent X-ray
powder diffraction).
[1] a) C. B. Aakerçy, N. R. Champness, C. Janiak, CrystEngComm 2010,
12, 22–43; b) M. M. Conn, J. Rebek Jr., Chem. Rev. 1997, 97, 1647–
1668; c) R. Custelcean, Chem. Commun. 2008, 295–307; d) G. R.
Desiraju, Angew. Chem. 2007, 119, 8492–8508; Angew. Chem. Int.
Ed. 2007, 46, 8342–8356; e) M. C. Etter, J. Phys. Chem. 1991, 95,
4601–4610; f) M. C. T. Fyfe, J. F. Stoddart, Acc. Chem. Res. 1997, 30,
393–401; g) D. S. Lawrence, T. Jiang, M. Levett, Chem. Rev. 1995,
95, 2229–2260; h) J.-M. Lehn, Angew. Chem. 1990, 102, 1347–1362;
Angew. Chem. Int. Ed. Engl. 1990, 29, 1304–1319; i) K. Merz, V. Va-
sylyeva, CrystEngComm 2010, 12, 3989–4002; j) D. Philp, J. F. Stod-
dart, Angew. Chem. 1996, 108, 1242–1286; Angew. Chem. Int. Ed.
Engl. 1996, 35, 1154–1196; k) L. J. Prins, D. N. Reinhoudt, P. Tim-
merman, Angew. Chem. 2001, 113, 2446–2492; Angew. Chem. Int.
Ed. 2001, 40, 2382–2426; l) C. M. Reddy, G. R. Krishna, S. Ghosh,
CrystEngComm 2010, 12, 2296–2314; m) G. M. Whitesides, J. P. Ma-
thias, C. T. Seto, Science 1991, 254, 1312–1319.
[18] S. Tragl, H.-J. Meyer, Z. Anorg. Allg. Chem. 2005, 631, 2300–2302.
[19] D. R. Miller, D. C. Swenson, E. G. Gillan, J. Am. Chem. Soc. 2004,
126, 5372–5373.
[20] a) C. Clauss, J. Wagler, M. Schwarz, A. Schwarzer, E. Kroke, Z.
Anorg. Allg. Chem. 2010, 636, 196–200; b) E. Horvath-Bordon, E.
Kroke, I. Svoboda, H. Fuess, R. Riedel, New J. Chem. 2005, 29,
693–699; c) S. J. Makowski, W. Schnick, Z. Anorg. Allg. Chem.
2009, 635, 2197–2202; d) A. Sattler, W. Schnick, Eur. J. Inorg.
Chem. 2009, 4972–4981.
[2] G. M. Whitesides, E. E. Simanek, J. P. Mathias, C. T. Seto, D. N.
Chin, M. Mammen, D. M. Gordon, Acc. Chem. Res. 1995, 28, 37–
44.
[3] a) G. Couderc, J. Hulliger, Chem. Soc. Rev. 2010, 39, 1545–1554;
b) P. H. Dinolfo, J. T. Hupp, Chem. Mater. 2001, 13, 3113–3125;
c) P. J. Langley, J. Hulliger, Chem. Soc. Rev. 1999, 28, 279–291.
[4] a) I. S. Choi, X. Li, E. E. Simanek, R. Akaba, G. M. Whitesides,
Chem. Mater. 1999, 11, 684–690; b) X. Li, D. N. Chin, G. M. White-
sides, J. Org. Chem. 1996, 61, 1779–1786; c) J. P. Mathias, E. E. Si-
manek, C. T. Seto, G. M. Whitesides, Angew. Chem. 1993, 105,
1848–1850; Angew. Chem. Int. Ed. Engl. 1993, 32, 1766–1769;
d) C. T. Seto, J. P. Mathias, G. M. Whitesides, J. Am. Chem. Soc.
1993, 115, 1321–1329; e) C. T. Seto, G. M. Whitesides, J. Am. Chem.
Soc. 1990, 112, 6409–6411; f) C. T. Seto, G. M. Whitesides, J. Am.
Chem. Soc. 1991, 113, 712–713.
[5] a) J. P. Mathias, C. T. Seto, E. E. Simanek, G. M. Whitesides, J. Am.
Chem. Soc. 1994, 116, 1725–1736; b) C. T. Seto, G. M. Whitesides, J.
Am. Chem. Soc. 1993, 115, 905–916.
[6] A. Ranganathan, V. R. Pedireddi, C. N. R. Rao, J. Am. Chem. Soc.
1999, 121, 1752–1753.
[21] S. J. Makowski, D. Gunzelmann, J. Senker, W. Schnick, Z. Anorg.
Allg. Chem. 2009, 635, 2434-2439.
[22] N. E. A. El-Gamel, L. Seyfarth, J. Wagler, H. Ehrenberg, M.
Schwarz, J. Senker, E. Kroke, Chem. Eur. J. 2007, 13, 1158–1173.
[23] a) Y. Ke, D. J. Collins, D. Sun, H.-C. Zhou, Inorg. Chem. 2006, 45,
1897–1899; b) T. Saplinova, V. Bakumov, T. Gmeiner, J. Wagler, M.
Schwarz, E. Kroke, Z. Anorg. Allg. Chem. 2009, 635, 2480–2487;
c) T. Saplinova, C. Lehnert, U. Bçhme, J. Wagler, E. Kroke, New J.
Chem. 2010, 34, 1893–1908; d) M. R. Schwarz, H. Ehrenberg, M. A.
Kloc, E. Kroke, Heterocycles 2006, 68, 2499–2507; e) A. Schwarzer,
E. Kroke, Chem. Commun. 2010, 46, 2829–2831; f) A. Schwarzer, E.
Kroke, New J. Chem. 2011, 35, 953–958; g) B. Traber, T. Oeser, R.
Gleiter, M. Goebel, R. Wortmann, Eur. J. Org. Chem. 2004, 4387–
4390.
[24] B. Jꢂrgens, E. Irran, J. Senker, P. Kroll, H. Mꢂller, W. Schnick, J.
Am. Chem. Soc. 2003, 125, 10288–10300.
[7] a) V. R. Pedireddi, S. Chatterjee, A. Ranganathan, C. N. R. Rao, J.
Am. Chem. Soc. 1997, 119, 10867–10868; b) A. Ranganathan, V. R.
Pedireddi, S. Chatterjee, C. N. R. Rao, J. Mater. Chem. 1999, 9,
2407–2411.
[8] a) D. J. Duchamp, R. E. Marsh, Acta Crystallogr. Sect. B 1969, 25, 5–
19; b) F. H. Herbstein, M. Kapon, G. M. Reisner, J. Inclusion
Phenom. 1987, 5, 211–217; c) S. V. Kolotuchin, E. E. Fenlon, S. R.
[25] A. Sattler, W. Schnick, Z. Anorg. Allg. Chem. 2006, 632, 238–242.
[26] A. Sattler, W. Schnick, Z. Anorg. Allg. Chem. 2006, 632, 1518–1523.
[27] a) L. Seyfarth, J. Sehnert, N. E. A. El-Gamel, W. Milius, E. Kroke, J.
Breu, J. Senker, J. Mol. Struct. 2008, 889, 217–228; b) J. Wagler,
N. E. A. El-Gamel, E. Kroke, Z. Naturforsch. B 2006, 61b, 975–978.
3256
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3248 – 3257

Formation of a Heptazine Framework
FULL PAPER
[28] a) N. E. A. El-Gamel, J. Wagler, E. Kroke, J. Mol. Struct. 2008, 888,
204–213; b) A. Sattler, M. R. Budde, W. Schnick, Z. Anorg. Allg.
Chem. 2009, 635, 1933–1939; c) A. Sattler, W. Schnick, Z. Anorg.
Allg. Chem. 2006, 632, 531–533; d) A. Sattler, W. Schnick, Z.
Anorg. Allg. Chem. 2008, 634, 1063–1066; e) A. Sattler, W. Schnick,
Z. Anorg. Allg. Chem. 2008, 634, 457–460; f) A. Sattler, W. Schnick,
Z. Anorg. Allg. Chem. 2010, 636, 2589–2594; g) A. Sattler, L. Sey-
farth, J. Senker, W. Schnick, Z. Anorg. Allg. Chem. 2005, 631, 2545–
2554.
[33] J. Eichhorn, S. Schlçgl, B. V. Lotsch, W. Schnick, W. M. Heckl, M.
Lackinger, CrystEngComm 2011, 13, 5559–5565.
[34] A. Sattler, Diploma Thesis, University of Munich (LMU), 2005.
[35] T. Steiner, Angew. Chem. 2002, 114, 50–80; Angew. Chem. Int. Ed.
2002, 41, 48–76.
[36] a) A. P. Cꢆtꢇ, A. I. Benin, N. W. Ockwig, M. OꢅKeeffe, A. J. Matzg-
er, O. M. Yaghi, Science 2005, 310, 1166–1670; b) A. P. Cꢆtꢇ, H. M.
El-Kaderi, H. Furukawa, J. R. Hunt, O. M. Yaghi, J. Am. Chem. Soc.
2007, 129, 12914–12915.
[29] E. Horvath-Bordon, E. Kroke, I. Svoboda, H. Fueß, R. Riedel, S.
Neeraj, A. K. Cheetham, Dalton Trans. 2004, 3900–3908.
[30] A. Sattler, S. Schçnberger, W. Schnick, Z. Anorg. Allg. Chem. 2010,
636, 476–482.
[31] B. V. Lotsch, W. Schnick, Chem. Eur. J. 2007, 13, 4956–4968.
[32] A. Sattler, S. Pagano, M. Zeuner, A. Zurawski, D. Gunzelmann, J.
Senker, K. Mꢂller-Buschbaum, W. Schnick, Chem. Eur. J. 2009, 15,
13161–13170.
[37] Z. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307–326.
[38] a) G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122;
b) G. M. Sheldrick, SHELXS-97, Program for the Solution of Crys-
tal Structures, Universitꢃt Gçttingen, 1997; c) G. M. Sheldrick,
SHELXL-97, Program for the Refinement of Crystal Structures,
Universitꢃt Gçttingen, 1997.
Received: November 11, 2011
Published online: February 7, 2012
Chem. Eur. J. 2012, 18, 3248 – 3257
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3257